AI & Agents
What a model or an agent is given and allowed to do: access to the code, the tools it may call, the retrieval behind its answers, and the guardrails around all of it.
Capabilities
Checking model input and output against content rules
Getting structured data out of incoming documents
Letting a coding agent read and act on your project or system
Checking an AI feature still behaves after a model or prompt change
Packaged knowledge that teaches a coding agent a framework
Building an agent into your own application
Replacing personal values in a prompt with placeholders
Calling a language model with tenant and credential handling solved
Prompt templates stored and versioned apart from the code
Grounding a model's answer in data retrieved at question time
Controlling what an agent may call, with an audit trail
Giving an AI agent access to the systems you run over MCP
Recent changes
All changes →Agent evals core coverage reassessed
Agent tooling core coverage reassessed
Document extraction coverage reassessed
LLM integration core coverage reassessed
Prompt management coverage reassessed
RAG on CDS core coverage reassessed